An Automatic Bottom Extracting Method for Side-Scan Sonar Image

The traditional side scan sonar (SSS) bottom tracking method is difficult to achieve ideal performance because the data is always influenced by complicated noise such as suspended solids, transmitted pulse, echoes from sea surface and ship wake flow. This will bring serious distortion and dislocation in the target image after SSS image processed by slant range correction. Based on the imaging mechanism and characteristic of side-scan sonar, this paper proposes the method of the last peak amplitude, the repair method of bottom extracting based on the trend of seafloor variation and the symmetry principle. Combining the Kalman filter and the characteristic and applicable object of the above methods, then gives an adaptive comprehensive bottom tracking and extracting method and the complete data processing flow. The proposed method achieved the function of tracking bottom automatically of SSS images obtained in an area of Yantai, where those SSS images were influenced by complicated marine noise and suspended solids. Compared with the external sounding data, the root mean deviation is ±0.17 m which shows the proposed method has high accuracy.